卷积过程:
将卷积过程迁移到图卷积中需要考虑两个问题:
(i) Determining the node sequences for which neighborhood graphs are created;决定围绕哪些结点创建邻接图(比如上图1中的结点1-4)
(ii) computing a normalization of neighborhood graphs, that is, a unique mapping from a graph representation into a vector space representation.得到图的表示之后,如何向量化
最早提出“基于时空的图神经网络”论文:Learning Convolutional Neural Networks for Graphs 阅读笔记
最新推荐文章于 2024-09-18 15:45:52 发布